블로그 이미지
Leeway is... the freedom that someone has to take the action they want to or to change their plans.
maetel

Notice

Recent Post

Recent Comment

Recent Trackback

Archive

calendar

1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31
  • total
  • today
  • yesterday

Category

By Bernard Kolman, Robert Edward Beck
Contributor Robert Edward Beck

Edition: 2, illustrated
Published by Academic Press, 1995
ISBN 012417910X, 9780124179103
449 pages



Prologue
: Introduction to Operations Research




Definitions

Operations research (OR) is a scientific method for providing a quantitative basis for decision making (that can be used in almost any field of endeavor).

http://en.wikipedia.org/wiki/Operations_research

The techniques of OR give a logical and systematic way of formulating a problem so that the tools of mathematics can be applied to find a solution (when the established way of doing things can be improved.

A central problem in OR is the optimal allocation of scarce resources (including raw materials, labor, capital, energy, and processing time).

eg. T. C. Koopmans and L. V. Kantorovich, the 1975 Nobel Prize awardees



Development

WWII, Great Britain
the United States Air Force
1947, George B. Dantzig, the simplex algorithm (for solving linear programming problems)
programmable digital computer (for solving large-scale linear programming problems)
1952, the National Bureau of Standards SEAC machine



Phases

> Steps
1. Problem definition and formulation
to clarify objectives in undertaking the study
to identify the decision alternatives
to consider the limitations, restrictions, and requirement of the various alternatives
2. Model construction
to develop the appropriate mathematical description of the problem
: limitations, restrictions, and requirements -> constraints
: the goal of the study -> quantified as to be maximized or minimized
: decision alternatives -> variables
3. Solution of the model
: the solution to the model need not be the solution to the original problem
4. Sensitivity analysis
to determine the sensitivity of the solution to changes in the inpuit data
: how the solution will vary with the variation in the input data
5. Model evalution
to determine whether the answers produced by the model are realistic, acceptable and capable of implementation
6. Implementation of the study



The Structure of Mathematical Models

There are two  types of mathematical models: deterministic and probabilistic.

A deterministic model will always yield the same set of output values for a given set of input values, whereas a probabilistic model will typically yield many different sets of output values according to some probability distribution.


Decision variables or unknowns
to describe an optimal allocation of the scarce resources represented by the model
Parameters
: inputs, not adjustable but exactly or approximately known
Constraints
: conditions that limit the values that the decision variables can assume
Objective function
to measures the effectiveness of the system as a function of the decision variables
: The decisin variables are to be determined so that the objective function will be optimized.



Mathematical Techniques

mathematical programming

Mathematical models call for finding values of the decision variables that maximize or minimize the objective function subject to a system of inequality and equality constraints.

Linear programming

Integer programming

Stochastic programming

Nonlinear programming

network flow analysis

standard techniques vs. heuristic techniques (improvised for the particular problem and without firm mathematical basis)


Journals

Computer and Information Systems Abstracts Journal
Computer Journal
Decision Sciences
European Journal of Operational Research
IEEE Transactions on Automatic Control
Interfaces
International Abstracts in Operations Research
Journal of Computer and System Sciences
Journal of Research of the National Bureau of Standards
Journal of the ACM
Journal of the Canadian Operational Research Society
Management Science (published by The Institute for Management Science - TIMS)
Mathematical Programming
Mathematics  in Operations Research
Naval Research Logisics (published by the Office of Naval Research - ONR)
Operational Research Quarterly
Operations Research (published by the Operations Research Society of America - ORSA)
Operations Research Letters
OR/MS Today
ORSA Journal on Computing
SIAM Journal on Computing
Transportation Science
posted by maetel
2008. 12. 29. 16:39 Computer Vision
MIT 18.06 Linear Algebra
lectured by Gilbert Strang

text: Introduction to Linear Algebra, 4th ed, Wellesley-Cambridge Press, Wellesley-Cambridge Press
official: http://math.mit.edu/linearalgebra/
google: http://books.google.co.kr/books?id=Gv4pCVyoUVYC

lecture http://web.mit.edu/18.06/www/
MIT 18.06 Linear Algebra, Spring 2005 @YouTube  videos @MIT
MIT 18.06 Linear Algebra, Spring 2010



Lecture 01: The Geometry of Linear Equations

n linear equations, n unknowns
1) Row picture -> 2 lines in x-y plane, 3 planes in x-y-z space, etc.
2) Column picture -> Linear combination of columns
3) Matrix form

Q1. Can I solve Ax=b for every b?
Q2. Do the linear combinations of the columns fill 3-D space?

"Ax is a combination of the columns of A."



Lecture 02: Elimination with Matrices

Elimination - success/failure
- augmented matrix
- pivot
Back-substitution
Elimination matrices
Matrix multiplication
- permutation matrix
- inverses




Lecture 03: Multiplication and Inverse Matrices

Matrix multiplication
Inverse of A  AB  A^t
Gauss-Jordan / Find A^-1



Lecture 04: Factorization into A=LU

Inverse of AB  A^T
Product of elimination matrices
A=LU (no row exchanges)



Lecture 05: Transpose, Permutations, Spaces R^n

section 2.7: PA = LU
section 3.1: Vector spaces and subspaces

"All the linear combinations of the columns of a matrix form a subspace, called column space C(A)."



Lecture 06: Column Space and Nullspace

Vector  spaces and subspaces
Column space of A: Solving Ax=b
Nullspace of A



Lecture 07: Solving Ax=0 : Pivot Variables, Special Solutions

Computing the nullspace (Ax=0)
Pivot Variables - free variables
Special Solutions - rref(A)=R


Lecture 08: Solving Ax = b: Row Reduced Form R

Complete solution of Ax = b
: x = x_p + x_n
Rank r
r = m : Solution exists
r = n : Soultion is unique


Lecture 09: Independence, Basis, and Dimension

Linear independence
Spanning a space
BASIS and Dimension


Lecture 10: The Four Fundamental Subspaces

(Correct an error in Lecture 9!)
Four Fundamental Subspaces (for matrix A)



Lecture 11: Matrix Spaces; Rank 1; Small World Graphs

Basis of new vector spaces
Rank one matrices
Small world graphs

http://en.wikipedia.org/wiki/Euclidean_subspace

http://en.wikipedia.org/wiki/Dimension_(vector_space)

http://en.wikipedia.org/wiki/Basis_(linear_algebra)

http://en.wikipedia.org/wiki/Rank_(linear_algebra)


Lecture 12: Graphs, Networks, Incidence Matrices

Graphs & Networks
Incidence matrices
Kirchhoff's Laws

http://en.wikipedia.org/wiki/Incidence_matrix
a matrix that shows the relationship between two classes of objects

입사 행렬(incidence matrix): 그래프 내의 모서리(edge)에 대한 정보를 담고 있는 2차원 배열. 그래프 내의 정점이 n개이면 nxn의 배열로 구성되며 배열의 각 원소는 해당 행과 열의 정점이 모서리로 연결되어 있는지를 0 또는 1로 나타낸다.  (출처 : [인터넷] 돌도끼 컴퓨터용어사전)


"Real matrices from genuine problems have structures."

"The null space tells us what combination of the columns to get zero."

"All the dependencies come from the loops."

"TREE is the name for a graph with no loop."


http://en.wikipedia.org/wiki/Planar_graph
Euler's formula states that if a finite, connected, planar graph is drawn in the plane without any edge intersections, and v is the number of vertices, e is the number of edges and f is the number of faces (regions bounded by edges, including the outer, infinitely-large region), then      v − e + f = 2.


Lecture 13: Quiz 1 Review

Emphasizes Chapter 3



null space ㅗ row space


"Null space contains all vectors ㅗ row space."







'Computer Vision' 카테고리의 다른 글

[유찬기] OpenCV & CV study guide  (0) 2009.01.09
.ai 파일 - 7cm*7cm grid  (0) 2009.01.05
Aleš Leonardis 강연 <Learning Hierarchical Representations of Object Categories>  (0) 2008.12.07
convex optimization  (0) 2008.11.28
papers  (0) 2008.11.14
posted by maetel
2008. 11. 5. 11:33 Method/Nature

An Introduction to the Conjugate Gradient Method Without the Agonizing Pain
Edition 1+1/4
Jonathan Richard Shewchuk
August 4, 1994
School of Computer Science, Carnegie Mellon University


http://en.wikipedia.org/wiki/Residual_(numerical_analysis)

http://en.wikipedia.org/wiki/Line_search

'Method > Nature' 카테고리의 다른 글

equation editor  (0) 2010.05.26
mathematical morphology  (0) 2009.02.14
MAT2110: 선형대수학 Linear Algebra  (0) 2008.09.05
Linear Algebra resources  (0) 2008.08.22
Superformula  (0) 2007.07.20
posted by maetel